Industry

Medical & Health Devices

Health devices often have to work where there is no internet, on hardware with little memory, and with results a clinician can trust. We build the firmware that captures the signal, the models that interpret it, the apps that show it, and the automated tests that prove each release still behaves.

Problems we are usually called for

  • Models that need a server or a heavy ML runtime the device cannot support
  • Noisy biosignals on small microcontrollers with no operating system
  • Firmware changes validated by hand on the bench, release after release
  • Clinical dictation and records that need structured, checked output

What we deliver

  • ✓On-device inference with no server dependency, including pure-Dart and quantised models
  • ✓Bare-metal acquisition firmware: multiplexed ADC, DMA, digital filtering
  • ✓Hardware-in-the-loop test suites and validation pipelines with coverage reporting
  • ✓Honest model evaluation: held-out test sets, per-split metrics and failure cases
  • ✓Companion apps on Flutter, Android and iOS, published under your account

Case studies in medical & health devices

Machine Learning Projects

HemoGuard: offline thalassemia screening on a phone

NUST, final-year project · 2025

A Flutter app that photographs a CBC blood report, reads the values with on-device OCR and screens for thalassemia with an XGBoost model re-implemented in Dart. No server, no internet, under 50 ms per prediction.

Read the case study →

Embedded Firmware

Automated firmware test suite for STM32F4 rehabilitation devices

Articares, Singapore · Jun 2025 – Aug 2025

A Python test suite and validation pipeline for STM32F4-based rehabilitation robots, with real-time performance dashboards; reached 95% test coverage.

Read the case study →

Embedded Firmware

EMG acquisition firmware on STM32F103 and 8051

RiseTech · Summer 2024

Firmware in C and assembly for EMG signal acquisition: multiplexed ADC sampling and a 50–500 Hz Butterworth band-pass filter, debugged with on-chip debuggers and logic analysis.

Read the case study →

AI Chatbots & Voice Agents

Physician-assistant AI pipeline

CareCloud (MTBC) · Aug 2025 – Dec 2025

A voice-to-structured-notes pipeline for physicians using LangChain, GPT-4, Whisper speech recognition and Gemini, plus a patient-profiling ML system deployed as rate-limited microservices.

Read the case study →

Relevant services

Machine Learning Projects

Custom models trained, evaluated and deployed properly.

Embedded Firmware

Microcontroller firmware for STM32, ESP32, nRF and similar.

Flutter Apps

One codebase for Android and iOS, built with Flutter.

AI Chatbots & Voice Agents

Custom chatbots and voice agents, fine-tuned on your data.